WO2015072612A1

WO2015072612A1 - Method for correcting typographical error by determination of correlation between character strings

Info

Publication number: WO2015072612A1
Application number: PCT/KR2013/010910
Authority: WO
Inventors: 김민철
Original assignee: 주식회사 큐키
Priority date: 2013-11-13
Filing date: 2013-11-28
Publication date: 2015-05-21
Also published as: US20160132471A1; KR101381101B1; US10755043B2

Abstract

The present invention relates to a method for correcting a typographical error, in which a user inputs a character string to be corrected and a character string determined as having a high correlation is substituted with the character string to be corrected and inputted by calculating a correlation between the character string to be corrected and the character strings previously inputted. The characteristic feature of the present invention lies in the correlation determining step. In the correlation determining step of the present invention, a plurality of typographical error corrections are defined, a typographical error correction score is assigned to each of the plurality of typographical error corrections, and the total score is calculated in a manner of adding up typographical error correction scores given to each of the typographical error corrections required to correct a previously inputted character string to a character string to be corrected. If the number of cases of typographical error corrections which can be selected for the correction is two or more, the score for the case having the lowest total score is used in determining the correlation. The correlation is determined by comparing a predetermined threshold value with the total score calculated in such a manner.

Description

How to correct typo by determining association between strings

The present invention relates to a method for correcting a typo by determining an association between character strings input from a mobile device such as a smartphone or a PC.

If you make a typo while typing characters (including English, Korean, numbers, special characters, etc.) on your PC or mobile device, the usual way to correct typos is to delete the backspace. Use the function keys to erase all characters entered after the typo, then type the correct character again, or move the cursor to the typo using the mouse (for PC) or finger touch (for mobile devices) I corrected a typo by deleting the characters and retyping the correct characters.

However, a method of correcting a typo using a delete function key such as a backspace has a problem in that not only a typo character but also all characters normally input between the typo character and the current cursor position are erased. In addition, the method of correcting a typo by moving the cursor to a typo character using the mouse is to release the mouse from the keyboard, move the cursor to the desired position, and move the cursor back to the current position after correcting the typo. There is a problem that the operation is cumbersome and time consuming. The same is true when the cursor is moved using a finger touch. In the case of a device having a small touch screen such as a mobile device, it is more difficult and cumbersome to move the cursor to the correct position because a finger touch must move the cursor to a desired position within the small screen.

In order to solve this problem, a so-called automatic correction function called an error correction method has appeared. This function is already commercially available and installed in various mobile devices. When the user enters the word A and the word A is determined to be a typo, the function automatically corrects the word B. The word B, which is automatically corrected, is already entered in the library and is not directly entered by the user.

As described above, the conventional automatic correction function sets the word that the system assumes to be entered by the user as a corrected input word instead of the word that the user inputs with the intention of correcting it. The wrong word is entered that is not. This is due in part to the limitations of words registered in the library. Because users frequently use new words or abbreviations on PCs or mobile devices, the problem caused by the limitations of these libraries becomes more serious when the library is not updated frequently. For this reason, many users are entering text while the device is equipped with automatic word correction, but this feature is turned off to disable it.

In order to supplement this problem of the conventional automatic correction function, the inventor invented a completely new error correction method. In the new typo correcting method devised by the present invention, when a user inputs a new string and issues a modification command, the new string is recognized as a desired target string for modification, and the new string is already inputted by comparing the input string with each other. Among the existing strings, a string highly related to the new string is determined as a candidate candidate for modification and replaced with a new string. In addition to the user's direct modification command, the system can automatically determine the user's intention to modify. This new typo correction method is registered in Korea Patent Registration No. 10-1294558.

In this new typo correcting method, the wrong input string is replaced with a new string newly input by the user with the intention of correcting, and thus, a typo correcting function corresponding to the user's intention can be provided as compared with the conventional automatic correction function. This method of correcting a typo is designed to replace the highly related string by judging the association between the newly entered modification target string with the intention of modification and the previously entered string. Therefore, the accuracy of the typo correction function is determined.

As described in the specification of Korean Patent No. 10-1294558, the association determination method is a conventional method of determining the association between a word actually input by a user and a word in a library in a conventional automatic correction function, such as in a keyboard. A method of determining association based on neighboring characters or characters whose input order is reversed may be used. However, in such a conventional method, the quantitative quantification of the correlation determination has not been properly performed in consideration of various typographical operations that may occur by the user. In addition, a typo input habit that differs from user to user is not considered quantitatively and accurately in determining association. Furthermore, according to the typographical modification method of Korean Patent No. 10-1294558, a completely new function is implemented to replace only the misspelled string with the correct string without retyping the entire word including the already entered typo. Has not considered any particularity of the association determination between the substrings, so a new association determination method is required for a more accurate implementation of the typos correction method of Korean Patent No. 10-1294558.

An object of the present invention is to provide a method for correcting a typo through a method of determining an association between character strings that enables the user to accurately select a character string to be modified according to a user's intention of modification among strings already input.

In addition, it is an object of the present invention to provide a quantitative quantification of the association judgment in comprehensive consideration of various typo actions that may occur by a user.

It is also an object of the present invention to quantitatively consider different typographical input habits for different users in determining association.

In addition, an object of the present invention is to implement an accurate association determination method considering the specificity of the association determination between substrings in the implementation of the function of replacing only the misspelled string with the correct string without retyping the entire word including the typo already entered. To provide.

The present invention provides an error correction method that can correct the error in accordance with the user's intention to correct. The typo correcting method of the present invention inputs a modification target string, determines the association between the modification target string and a previously input string, and replaces a string having a high association with the modification target string among the previously input strings with the modification target string. On the premise that The present invention is characterized in that the method of determining the association between the target string to be corrected and the character string previously input in the error correction method.

In the method for determining an association between a modification target string and a character string previously inputted according to the present invention, first, a plurality of typo correction operations required to modify the character string input before the modification target string into the modification target string are defined. A predetermined score is assigned to each of a plurality of typo correcting operations defined. The total score is calculated in such a manner that the points assigned thereto are added up each time a modification operation defined in the modification of the string input before the modification target string into the modification target string occurs. When the number of possible typo correcting operations for correction is plural, the sum of scores is minimized. The best case for correcting the character string to the modification target string is that the score at this time, that is, the minimum score is related. It is considered in the judgment. Scores added in this manner are compared with a predetermined threshold value, and when the sum score is less than or equal to the threshold value, it is determined that the corresponding character string has a high correlation with the modification target character string. In comparing the sum score with the threshold value, the length of the string to be compared may be considered. If the string is long, it is likely to contain a relatively large number of typos. Nevertheless, if the thresholds remain the same, the summation score will often exceed the threshold. Therefore, if the length of the string is long, the threshold can be increased, and if the length of the string is short, the threshold can be implemented. In addition to adjusting the threshold value according to the length of the string to be compared in this manner, a method of considering the length of the string to be compared to the sum score of the points given to each typo correcting operation, for example, (sum of points / characters of the string to be compared) The number obtained through the calculation of number) can also be implemented by performing a comparison with a predetermined threshold value.

The typo correcting operation used in the association determination method according to the present invention includes replacing a character in a string with a character of a neighboring keyboard, replacing a character in the string with a character of a non-neighboring keyboard, 2 consecutively included in the string. The operations of exchanging the positions of two characters, inserting characters into a string, and deleting characters from the string. As described above, each operation is given a predetermined score. The points given to each typo correcting operation may be the same or different.

In addition, in the method for correcting a typo according to the present invention, it is possible to replace only a misspelled string with a correct string without retyping a whole word including a typo already entered. Weight the characters. The weighting method is determined by adding the scores to each typo correcting operation described above, and then adding a predetermined value if the first character of the string compared with the target string is not equal to or adjacent to the first character of the target string. It is implemented in such a manner that an additional score is given and a predetermined additional score is given if the last character of the string compared with the modification target string is not equal to or adjacent to the last character of the modification target string.

According to the present invention, the score given to each typo correcting operation may be configured to be adjusted according to the frequency of occurrence of the typo correcting operation. For example, if the user frequently inputs the position of two characters consecutively included in the character string, the character string having the same typo pattern is corrected by gradually adjusting the score given to the typo correcting operation. It will gradually increase the likelihood that it is highly related to the target string. In this way, user customization is achieved.

According to the present invention, a score given to each typo correcting operation is generally assigned to the most versatile value by a user survey, but can be set directly or modified by the user.

According to the present invention, since the quantitative quantification of the association determination is provided in comprehensively considering various typoss that may occur by the user, the accuracy of the association determination is greatly improved.

In addition, according to the present invention, the score given to each typo correcting operation can be adjusted according to the frequency of occurrence of the typo correcting operation, thereby providing an intelligent association determination method adapted to a typo input habit of the user. This also contributes greatly to improving the accuracy of the association judgment.

In addition, according to the present invention, in the associative determination method, weights are given to the first and last characters of the two strings to be compared, thereby replacing only the misspelled string with the correct string without retyping the entire word including the typo. The approach can be implemented more accurately.

In addition, according to the present invention, since the user can directly set and correct the score given to each typo correcting operation, it is possible to reflect the user's intention in the association determination in advance.

1 illustrates an embodiment of correcting a typo in a character string input by a user's modify command.

2 illustrates a typo correcting operation of correcting a string in which some characters are incorrectly input as characters of a neighboring keyboard to a correct string.

3 illustrates a typo correcting operation of correcting a string in which some characters are incorrectly inputted as characters of a keyboard that is not adjacent to the correct string.

Fig. 4 illustrates a typo correcting operation in which the sequence of two consecutive characters is reversed to correct the input string to the correct string.

5 illustrates a typo correcting operation of correcting a string in which unnecessary characters are added to a correct string.

Fig. 6 shows a typo correcting operation of correcting a character string missing a necessary character to a correct character string.

FIG. 7 illustrates a typo correcting operation in which only a portion of a string in which a typo occurs is inputted as a modification target string to correct a typo.

8 illustrates an example of a setting screen configured to allow a user to set a typo correcting operation score.

FIG. 1 illustrates an embodiment of correcting a typo included in character strings input by a user's correction command with a newly input correction target character string. In FIG. 1, the user intends to input Happy Easter everyone, but actually inputs Hsppy Easter wveryone. The user first finds an error in the wveryone and puts a space after Hsppy Easter wveryone to fix it. After that, give a modify command. The modification command may be implemented in various forms. A modification command button may be separately formed in the device, or a touch action for gently shaking the mobile device or moving the keyboard area up, down, left, and right may be recognized as a modification command. If the modification command is issued after everyone's input in any way, the system recognizes everyone as the modification target string. After that, the association between the modification target string everyone and the previously entered strings is calculated. Since wveryone is highly related to everyone who is the target string to be modified, wveryone is replaced by everyone, and Hsppy Easter wveryone everyone is modified to Hsppy Easter everyone. A detailed association determination method will be described later. In this state, the user finds an additional typo in Hsppy, puts a space in everyone, enters the correct string happy, and issues the corrective command again. Then, happy is recognized as the modification target string, and the association with the strings already entered is calculated. As a result of the calculation, Hsppy is associated with happy, Hsppy is replaced by happy, and in the process, h is capitalized as happy is recognized as the first word of the sentence, and Hsppy Easter everyone is input as Happy Easter everyone. This completes the modification process. It is up to the system to determine whether the target string is modified against a range of previously entered strings, for example, against 10 previously entered strings, 100 strings, or strings throughout the document. It can be set freely according to the environment.

2 through 6 illustrate various types of typographical errors and corresponding typo correcting operations.

Fig. 2 shows a case where a part of the characters in the character string is wrongly input as the characters of the neighboring keyboard. In FIG. 2, the user intended everyone's input, but actually entered is wvwryone. If the user finds an error in wvwryone, enters everyone to correct it, then issues a correction command, everyone is recognized as the target string to be corrected and the association is calculated with the strings already entered. At this time, everyone and wveryone differ only in whether the first and third characters are e or w. Since e and w are input keys placed adjacent to each other on the keyboard, it is highly likely that the input e must have been entered incorrectly as w. Therefore, wvwryone is selected as a candidate string for modification that is highly related to everyone, and is modified to everyone, the target string for modification.

Fig. 3 shows a case where a part of the characters in the character string is wrongly input as the characters of the non-neighboring keyboard. In Fig. 3, the user intended everyone's input, but actually the input is everione. If you compare the string you entered incorrectly (everione) with the actual intended string (everyone), you can see that the two strings differ only in whether the fifth character is i or y. However, when it is determined that the characters are differently related to the two strings as shown in FIG. 2, only when the characters are adjacent to each other, the string is highly related. In FIG. 3, since the fifth characters i and y do not neighbor each other, everione It cannot be selected as a string that is highly related to the target string to be modified, and typos are not corrected.

The present invention defines various typo correcting actions and assigns a typo correcting operation score to each of these typo correcting actions so that even when characters of non-neighbory keyboards are incorrectly inputted as shown in FIG. Compare the sum of the typo correction action scores required to correct the already inputted string with the newly entered modification target string to determine whether the already inputted string is a string that is highly related to the newly entered modification target string. It is composed. Specifically, in the embodiment of the present invention, when the sum of the typo correcting operation scores is less than or equal to a predetermined threshold value, the two strings are configured to determine that they correspond to a string having a high correlation with each other. For example, in the present embodiment, as shown in Figs. 2 and 3, each typo correcting operation score is assigned to all of the operations for correcting the characters in the character string. However, in the case of the adjacent keyboard as shown in FIG. 2, a typo is more likely to occur than in the case of the non-neighbored keyboard shown in FIG. Therefore, in order to reflect this possibility in the correlation determination, when correcting the characters of the non-neighboring keyboard as shown in FIG. 3, it may be preferable that the error correction operation score is higher than the score given to the error correction operation of FIG. 2. In this embodiment, five points are given to the typo correcting operation of FIG. 2, and ten points are assigned to the typo correcting operation of FIG. In addition, in the present embodiment, a reference value (an error correction operation score threshold) for comparing such a mistake correction operation score is set to 15 points.

According to this association determination method, in order to modify wvwryone to everyone as the modification target string in FIG. 2, an operation of modifying the first and third letters w to the letters e of the neighboring keyboard is required. Since five points are set in the operation of modifying the first character w to e and five points are set in the operation of modifying the third character w to e, 10 points, the sum of these points, are a typo correcting operation score. Here, since the 10 typo correcting operation score is smaller than the 15 typo correcting threshold, the two strings are determined to be highly related strings, and wvwryone is selected as the candidate string for modification of everyone, the correct target string.

In addition, in order to correct everione everyone in FIG. 3, a single typo correcting operation of modifying the fourth character i to y is required. At this time, since i and y are characters of keyboards that do not neighbor each other, the typo correcting operation score of 10 points is assigned to the typo correcting operation. Since the 10 typo correcting operation score is smaller than the 15 typo correcting threshold, the two strings are judged to be highly related strings, and everione is selected as the candidate string for modification of everyone, the correct target string.

4 illustrates a case where two consecutive characters entered in a character string are inputted in reverse order. In FIG. 4, the user intends to input believe, but the input is actually beleive. In this case, the simplest action for correcting a typo is to swap the positions of the fourth character e and the fifth character i. However, to correct beleive to believe according to the typo correcting operation described with reference to FIGS. 2 and 3, the fourth character e must be changed to i and the fifth character i must be changed to e. A total of 20 typo correcting action scores are given to change twice, and since 20 typo correcting action scores are larger than the threshold value of 15 points, beleive is determined to be a string that is not related to believe, the correct target string. Will not be modified. In fact, when the sequence of letters is reversed and inputted in this way, characters are frequently generated during a high speed input process. Nevertheless, if the association of the character string is determined only by the association determination method described with reference to FIGS. 2 and 3, a typographical error that may occur frequently as shown in FIG. 4 may not be corrected.

In this embodiment, in order to solve such a problem, the operation of exchanging the sequence of two consecutive characters inputted into the character string is defined as a separate typo correcting operation, and the operation is given a typo correcting operation score of 8 points. . In the example shown in Fig. 4, the beleive can be modified to believe as a modification target string by an operation of exchanging positions of the fourth character e and the fifth character i, so that the beleive has an error correction operation score of eight points. Since the error correction operation score of 8 points is smaller than the threshold value of 15, the beleive is determined to be a string highly related to the believe target string believe and is corrected to believe.

5 and 6 illustrate a case where a typo occurs because an unnecessary character is inserted into an input character string or a required character is missing. Figure 5 is intended to enter the god, but in reality the case is inserted into the good and one more unnecessary o is inserted between g and d, Figure 6 is intended to enter good, but unintentionally o missing one input to god If it is. In fact, these errors can easily occur during the fast text entry process. In this case, an incorrectly written string may be modified through a corrective operation of deleting an unnecessaryly inserted o (in case of FIG. 5) or adding an unintentionally missing o (in case of FIG. 6). However, the typo correcting operation shown in Figs. 2 to 4 ('replacement of wrong characters' or 'position exchange between neighboring characters') is basically associated with a string of the same length as the newly input correction target string. Therefore, when an unnecessary character is inserted into a character string as shown in FIG. 5 or a necessary character is missing from the character string as shown in FIG. 6, a typographic error cannot be determined as a character string associated with the target string to be modified.

In order to solve such a problem, in the present embodiment, an operation of deleting a character from a string or adding a character to the string is defined as a separate typo correcting operation, and the operation is given a typo correcting operation score of 10 points.

In the example shown in Figure 5, god can be modified to good by adding one o between the first character g and the last character d, and in Figure 6 good is the first character g and the last character Can be modified to god by deleting one of the two o's entered between d. Accordingly, the correction operation shown in FIGS. 5 and 6 is given a typo correction operation score of 10 points, respectively. Here, since the typo correcting operation score of 10 points is smaller than the threshold value of 15 points, these strings are determined to be strings highly related to the target string of modifications and are selected as candidate strings for modification.

As described above, the present invention defines (i) a plurality of typo correcting operations to cover various typo types, (ii) assigns a typo correcting operation score to each of these typo correcting operations, and (iii) Calculate the minimum value of the total typo correcting action score required to correct the already inputted string with the newly input corrected target string, and (iv) compare the sum of the typo correcting action scores with a predetermined threshold value. Since it is configured to select a candidate string for modification having a high correlation with the string, it is possible to provide an error correction method that can correct the typo by more accurately reflecting the user's intention to modify.

FIG. 7 illustrates a case where a user finds an error in an input string, and then corrects a typo by inputting only a partial string included in the entire string as a modification target string. In FIG. 7, the user intends to input remember, but the input is actually renember. At this time, in order to correct a typo in a fast and economical manner, it is preferable to correct a typo by inputting only a partial string including a character in which a typo occurs as a target string, rather than newly inputting a correct entire string as a target string of correction. That is, in order to economically correct a typo in FIG. 7, it is preferable to input only the partial string including the n (third character), which is an incorrectly input character, as a correct target string rather than newly remembering the entire string as the modification target string. (E.g. rem, eme, mem, ...). As described above, when a user wants to correct a typo by inputting only some strings as a target string of correction, a user often enters a target string of characters from an incorrectly input character or a character string that is incorrectly input as a target string of correction. (Eg, enter mem or rem in the embodiment of FIG. 7). Therefore, when the typo correcting operation is performed at the beginning or end of the character string, it is necessary to give a weight to the typo correcting operation. In the present embodiment, when a typo correcting operation is performed on a character at the beginning or end of a string, when the character before and after the change is a 'character of a non-neighboring keyboard' that is less likely to cause a typo, the typo correcting score is 5 points. The addition point is added. According to this method, in the case of modifying the beginning or ending character of the already inputted string to modify the already inputted string into the modification target string, (i) the characters before and after the modification are not 'neighbors' which are less likely to cause a typo. 'Character of the keyboard', additional points are added to the typo correcting action score, and these strings are easily excluded from the candidate string for modification, and (ii) in the case of 'characters of the neighboring keyboard' that are more likely to make a typo. The probability of being retained as the candidate string for modification can be increased.

Hereinafter, a typo correcting operation according to the present invention will be described in detail with one example. 7 illustrates a case in which the user intends to input remember, but is actually incorrectly inputted as renember. The user first finds an error in renember, then puts a space after renember and enters mem to fix it. After that, give a modify command. When a modification command is issued, the system recognizes mem as the modification target string, and the association between the modification target string and mem and the string previously entered is calculated.

First, the association calculation is started on the string [ren, ene, nem, emb, mbe, ber] of the same length as mem, the newly modified target string. The typo correcting operation score for each string is as follows. As shown in the table below, only nem has a typo correcting operation score below the typo correcting operation score threshold (15 points) among the strings having the same length as the correcting target string. Therefore, only nem is selected as the candidate string for modification.

Table 1

String to compare	Modification goal string	Typo Correction Behavior Score
ren	mem	(1) Modify the first character to the character of a non-neighboring keyboard: 10 points (2) Modify the third character to the character of a neighboring keyboard: 5 points Point Correction: 20 points
ene	mem	(1) Modify the first character to the character of the non-neighbors keyboard: 10 points (2) Modify the second character to the character of the non-neighbors keyboard: 10 points Point (4) Correct the start character to the character of a non-neighboring keyboard: 5 additional points (5) Modify the end character to the character of the non-neighbored keyboard: 5 additional points Correct typos
nem	mem	(1) Correct the first letter to the letters of the neighboring keyboard: 5 typo Corrective action score: 5 points
emb	mem	(1) Modify the first character to the character of the non-neighbors keyboard: 10 points (2) Modify the second character to the character of the non-neighbors keyboard: 10 points Point (4) Correct the start character to the character of a non-neighboring keyboard: 5 additional points (5) Modify the end character to the character of the non-neighbored keyboard: 5 additional points Correct typos
mbe	mem	(1) Modify the second character to the character of a non-neighboring keyboard: 10 points (2) Modify the third character to the character of a non-neighboring keyboard: 10 points 5 typo Correction score: 25
ber	mem	(1) Modify the first letter to the text of a non-neighboring keyboard: 10 points (2) Modify the third letter to the text of a non-neighboring keyboard: 10 points Correction of 5 (4) end characters to characters of non-neighboring keyboards: 5 additional points Correction of typos Correction score: 30

Next, judging the methods of correcting a typo by the above-mentioned 'delete unnecessary characters' or 'add necessary characters' based on the target string and a string having a different length (for example, re, rene, ...). do. In this example, there is no correction operation having a typo correction operation score of 15 or less among the correction operations by this method. As a result, in this example, only nem is determined to be a string highly related to mem, nem is replaced with mem, and renember is modified to remember. As described above, the present invention does not compare only the strings having the same length as the modification target string, but also compares the strings having different lengths by one or more characters, thereby improving the accuracy of the association judgment.

On the other hand, the same user generally generates a similar type of typo repeatedly. Therefore, in order to allow the user to feel more satisfied with the typo correction function, the typo correction operation that is generated in the device is memorized, and the typos correction operation score is automatically adjusted according to the frequency of occurrence of the typo correction operation to be customized for the user. It is desirable to. According to this configuration, a low operation score is given to a typo correcting operation that is frequently generated by the user, and a high operation score is given to a typo correcting operation that is not frequently generated by the user. The character string can be easily selected as the candidate character string to be modified, and the user can be sufficiently satisfied with the error correction function.

In addition, it is preferable to configure a typo correcting operation score that is assigned to each typo correcting operation so that a user can directly set and correct it. 8 shows an example of a 'setting' screen configured to allow a user to set and correct a typo correcting operation score. The user may adjust a typo correcting operation score for each typo correcting operation to his / her preference by using a finger touch or the like on the setting screen as shown in FIG. 8. According to such a configuration, the user can set a method of determining the most appropriate association, so that the user can feel more satisfied with the error correction function.

The above embodiments are presented for illustrative purposes only to illustrate the present invention, but do not limit or limit the present invention. Those skilled in the art to which the present invention is applied can make various changes and modifications to the present invention without departing from the scope of the present invention.

Claims

In a typo correction method, a user inputs a modification target string, calculates an association between the modification target string and previously input strings, and substitutes the character string determined to have a high correlation with the modification target string.

In the association determination step

Multiple typo correcting actions are defined,

A typo correcting operation score is assigned to each of the plurality of typo correcting operations,

The total of the scores is calculated in such a manner that the typo correcting action scores assigned thereto are added to each typo correcting action required to correct the previously inputted string into the corrected target string. In this case, the score when the sum of the scores is the minimum is used to determine the association.

The association determination is made by comparing the sum of the calculated scores with a predetermined threshold.
The method of claim 1, wherein the plurality of typo correcting operations defined above include: replacing characters in a string with characters of a neighboring keyboard, replacing characters in a string with characters of a non-neighboring keyboard, and continuously described in the character string. A method for correcting a typo, comprising: exchanging positions of two characters with each other, inserting a character into a string, and deleting a character from the string.
The method of claim 1 or 2, wherein after calculating the total sum of scores by adding the scores assigned to each typo correcting operation in the associative determination step, the first character of the string to be compared with the modification target string is determined. If the first character is not equal to or adjacent to the first character, a certain additional score is given. If the last character of the string compared to the target character string is not equal to or adjacent to the last character of the target character string, the additional additional score is given. How to fix, typo correction.
The method of claim 1 or 2, wherein the error correction operation score is configured to be automatically adjusted according to the frequency of occurrence of the error correction operation.
The method of claim 1 or 2, wherein the error correction operation score is configured to be set or corrected by a user.
The method of claim 1 or 2, wherein in comparing the sum of the calculated scores with a predetermined threshold, the length of the string to be compared is adjusted by adjusting the threshold value or adjusting the sum of the scores according to the length of the string to be compared. To consider,

The method of correcting a typo, by comparing the sum of the calculated scores with a predetermined threshold.